New AI Tool Can Predict Cancer Survival by a Single Patient Selfie, Study Suggests
System can estimate a person’s biological age and predict chances of cancer survival from a single photograph
TL;DR
- An AI tool has been developed to predict cancer patient survival and biological age from a single facial photograph.
- The system, built by Mass General Brigham, was trained on over 40 million facial photographs.
- Two complementary tools, FAHR-FaceAge and FAHR-FaceSurvival, estimate biological age and predict survival directly, respectively.
- FAHR-FaceAge analyzes creases, forehead, and temples, while FAHR-FaceSurvival focuses on the area below the eyes and around the nasal bridge.
- Patients appearing five or more years older had a higher risk of dying, while those appearing younger had a lower risk.
- The AI's ability to learn from a large dataset provides a scalable approach to developing facial health biomarkers.
- Lifestyle factors like heavy alcohol use, drug use, and smoking were linked to appearing biologically older.
- Researchers caution that further research is necessary before the tool can be widely implemented.